The Second International Conference on Electrical ...

4 downloads 185628 Views 6MB Size Report
The Proceedings of Second International Conference on Electrical and Electronics Engineering, Clean Energy and Green Computing, ...... with an immediate call button, the location map, ..... while 48% of patients use the iPhone operating.
Conference Title

The Second International Conference on Electrical and Electronics Engineering, Clean Energy and Green Computing (EEECEGC2015) Conference Dates

May 26-28, 2015 Conference Venue

Mevlana University, Konya, Turkey ISBN

978-1-941968-12-3 ©2015 SDIWC

Published by

The Society of Digital Information and Wireless Communications (SDIWC) Wilmington, New Castle, DE 19801, USA www.sdiwc.net

Table of Contents Design and Analysis of Dual-Loop VHF Circularly Polarized Antenna for Low Earth Orbit Satellites Ground Stations.................1 Propagation Model for the 900 MHz Almadar Aljadid Mobile Network at Tripoli Area Using Linear Regression Method.........................................................................................................................................................................................5 A Survey on Some Currently Existing Intrusion Detection Systems for Mobile Ad Hoc Networks.............................................12 Harmonics Mitigation and Power Quality Improvement in Grid Connected Wind Power Plant Using STATCOM..........................................................................................................................................................................19 Proposed Optimization Technique for Maximization of Throughput under Using Different Multicarrier Systems in Cognitive Radio Networks.........................................................................................................................................25 A New Clustering Technique for Energy Efficient Wireless Sensor Networks with Static Base Stations using SVD and DWT..............................................................................................................................................................................34 Field-Oriented Control of DFIG in Wind Turbine.........................................................................................................................44 Compact Monopole Antennas for Ultra Wideband: Designs and Synthesis in the Band of 10-45 GHz......................................50 Consumers Attitude towards the Use of Mobile Health Apps: An Empirical Review.................................................................56

The Proceedings of Second International Conference on Electrical and Electronics Engineering, Clean Energy and Green Computing, Konya, Turkey, 2015

Design and Analysis of Dual-Loop VHF Circularly Polarized Antenna for Low Earth Orbit Satellites Ground Stations Rafael Haag and Georgia Andrade Tomaszewski State University of Rio Grande do Sul Av. Bento Goncalves, 8855, Porto Alegre, RS, 91540-000, BRAZIL [email protected] ABSTRACT The State University of Rio Grande do Sul is one of the partners to develop a National Satellite Network Stations for remote sensor data gathering and communication in Universities and Schools. The research has focused on a Low Earth Orbit (LEO) CubeSat system, with support from the Brazilian federal government’s and Brazilian Space Research Program. Our objective is to describe the implementation of a compact, low-cost antenna solution for the National Satellite Network, operating in the 145-146 MHz band. This paper proposed a single feeding port, low profile VHF omnidirectional circularly polarized (CP) antenna. The simplicity of the design coupled with broad impedance bandwidth and axial ratio make it suitable for satellite communication terminals.

KEYWORDS VHF antenna, Wire antenna, Circular polarization, Axial ratio, Radiation pattern.

1 INTRODUCTION Circularly Polarized (CP) antennas offer a distinct advantage over the Linearly Polarized (LP) antennas for satellite applications. The main advantage of CP versus linear polarization is that CP eliminates polarization mismatch losses caused by Faraday's rotation and varies the squint angle of polarization vectors between stations on the Earth. Conventional approach in designing CP planar loop/dipole antennas uses four or two separate antenna elements fed by their own feeding networks usually containing phase shifters or 90-degree hybrids,

ISBN: 978-1-941968-12-3 ©2015 SDIWC

which contributes to the system design complexity [1]. It is well known that a one-wavelength perimeter square loop antenna radiates linearly polarized waves [2]. The reason for the linear polarization production is due to the in-phase current distribution along its two radiating sides. A square loop antenna can also radiate a CP wave if a gap is introduced on the loop [3, 4, 5, 6]. The CP radiation is due to the travelling-wave current distribution that becomes excited along the loop. The sense of circular polarization can be switched from left-hand or right-hand, by altering the gap positions [7]. In this article, a simple CP dual-loop planar antenna is proposed. Theoretically this antenna can generate nearly ideally circular polarization at broadside (perpendicular to the loop plane). The structure of the proposed antenna is simple and can be fabricated using thin aluminum tube/wire (light-weight), which leads to the advantages such as low cost and low profile. The design processes of the proposed antenna are explicitly shown, and parametric studies through simulation based on a method of moment (MoM) were carried out. Typical experimental measurement was also conducted and compared to the simulated ones. 2 ANTENNA DESCRIPTION AND DESIGN Figure 1 shows the geometry of the proposed CP dual-loop. The antenna is designed to resonate in the 146 MHz (λ0=2.05m), the

1

The Proceedings of Second International Conference on Electrical and Electronics Engineering, Clean Energy and Green Computing, Konya, Turkey, 2015

Cubesat's VHF downlink band. In Figure 1, the radiating elements are metallic wire elements (aluminum) having a rectangular shape (10mm X 2mm). Length of the gaps section (g1 and g2) is optimized in Numerical Electromagnetics Code, NEC-2 [8] for the best axial ratio (AR). The optimized value for g1 and g2 is ≈0.01 λ0. The square side length L1 and L2 is ≈0.25 λ0, or 510mm. The length of the fp (gap of feed point) is about 10mm. The antenna requires a transformer circuit to realize a balanced feed. A simple unbalanced to balanced 4:1 impedance transformer made of coaxial cable was used.

as input impedance, height above a flat typical ground, electric field amplitudes/phases, and axial ratio (AR) have been investigated. Figure 2 shows the simulated and measured input impedances of the proposed dual-loop CP antenna. The Figure 3 shows the reactance simulated and measured. Good agreement was achieved between the simulated and measured results.

Figure 2. Measured and simulated input impedance.

Figure 1. Geometry of the proposed dual-loop CP antenna.

3 RESULTS AND DISCUSSIONS The goal of this study is to develop a planar dual-loop antenna with good CP radiation and good impedance match at the frequency of 146 MHz. In order to easily evaluate the circumference (length of L1 and L2) of the dual square-loop related to the desirable operational frequency, several simulations have been done by NEC-2 simulator. The influence of various parameters on the antenna performances such

ISBN: 978-1-941968-12-3 ©2015 SDIWC

Figure 3. Measured and simulated input reactance.

2

The Proceedings of Second International Conference on Electrical and Electronics Engineering, Clean Energy and Green Computing, Konya, Turkey, 2015

The measured return loss response of the antenna versus frequency is shown in Figure 4.

Figure 5. Simulated axial ratio with respect to the elevation in the xz plane. Figure 4. Measured return loss against frequency for the proposed antenna.

The antenna height was 5 meters above flat and unobstructed ground during the measured of return loss in Figure 4. The measured impedance bandwidth for a return loss more than -10 dB was approximately 20 MHz (135–155 MHz). From the above results, it has proven that the bandwidth of this antenna is more than adequate for use in the Cubesats VHF segment (145-146 MHz). The Figure 5 shows the simulated axial ratio with respect to the elevation in the xz plane. The discrepancy in the symmetry of the axial ratio for elevation more than 15º in xz plane in the right part of Figure 5 is explained by the presence of the gaps on the loop corners and the effect of the edges on the current distribution. The elevation-angle ranges, when AR values are less than 3dB, are -50 to 65 degrees at 146 MHz.

ISBN: 978-1-941968-12-3 ©2015 SDIWC

Figure 6 presents the simulated antenna gain for the proposed dual-loop antenna, and steady gain variation of no more than 0.2 dB was observed between 140 and 150 MHz. At 146 MHz, gain of 7.3 dBic was attained by the proposed CP loop antenna.

Figure 6. Simulated antenna gain for the antenna studied in Figure 1.

3

The Proceedings of Second International Conference on Electrical and Electronics Engineering, Clean Energy and Green Computing, Konya, Turkey, 2015

4 CONCLUSIONS A simple planar omnidirectional CP square dual-loop antenna for VHF LEO satellite application has been successfully studied and implemented. The impedance bandwidth were 20 MHz (135–155 MHz). The gain at 146 MHz was 7.3 dBic, and steady gain variation of no more than 0.2 dB was observed between 140 and 150 MHz. The proposed antenna has good broadside CP radiation patterns at least covering the range of 115 degrees in elevation. Based on the above, we conclude that the proposed antenna has excellent performance and is highly recommended for LEO satellites ground stations.

polarization and impedance matching,” IEEE Transactions on Antennas and Propagation, vol. 52, pp. 551–554, 2004. [9] R. L. Li. V. F. Fusco, and R. Cahill, “Pattern shaping using a reactively loaded wire [10] loop antenna,” IEE Proc.-Microwave, Antennas Propag., vol. 148, pp. 203-208, June [11] 2001. [12] G. Burke and A. Poggio. Numerical electromagnetics code - method of moments. Technical Report UCID-18834, Lawrence Livermore National Laboratory, CA, 1981.

ACKNOWLEDGMENT This work was supported by the National Counsel of Technological and Scientific Development (CNPq) under Grant 420277/2013-7.

REFERENCES [1] H. Nakano, K. Nogami, S. Arai, H. Mimaki, and J. Yamauchi, “A spiral antenna [2] backed by a conducting plane reflector,” IEEE Trans. Antennas Propagat., vol. AP-34, [3] pp. 791-796, June 1986. [4] C. A. Balanis, Antenna Theory. New York: Harper and Row, 1982, pp. 231-275. [5] H. Morishita and K. Hirasawa, “Wideband circularly-polarized loop antenna,” IEEE Antennas and Propagation Society International, Symposium Digest, vol. 2, pp. 1286–1289, Seattle, WA, 1994. [6] R. L. Li, V. F. Fusco, and H. Nakano, “Circularly polarized open-loop antenna,” IEEE Transactions on Antennas and Propagation, vol. 51, no. 9, pp. 2475–2477, 2003. [7] R. Li, G. De Jean, J. Laskar, and M. M. Tentzeris, “Investigation of circularly polarized loop antennas with a parasitic element for bandwidth enhancement,” IEEE Transactions on Antennas and Propagation, vol. 53, no. 12, pp. 3930–3939, 2005. [8] M. Sumi, K. Hirasawa, and S. Shi, “Two rectangular loops fed in series for broadband circular

ISBN: 978-1-941968-12-3 ©2015 SDIWC

4

The Proceedings of Second International Conference on Electrical and Electronics Engineering, Clean Energy and Green Computing, Konya, Turkey, 2015

Propagation Model For the 900 MHz Almadar Aljadid Mobile Network at Tripoli Area Using Linear Regression Method Eng. Rabie Abboud Radio Network Optimization Department Almadar Aljadid Mobile Company [email protected] ABSTRACT The accuracy of the predicted Received Signal Strength (RSS) is the key issue in the mobile network design. In the last decades many models invented to predict the RSS in specific locations. One of the most famous models is Hata model, which is developed originally based on measurements done in Tokyo – Japan. The objective of this paper is to modify Hata model at frequency band 900 MHz to match the environments of Tripoli area, the capital of Libya. The modifications was based on real measurements of the RSS which were conducted in the concerned area. The modified model is expected to be used during the designing phase of the mobile networks in the future. The measurements is collected in five types of areas (Dense Urban, Urban, Dense Suburban, Suburban and Rural). The regression method is used to modify the Hata model parameters to match the real measured values of RSS within an acceptable range of accuracy. The RMSE (root mean square error) was calculated and used as an assessment factor for the modified module. The modified module gives 4.7 to 23.6 dB improvement in the accuracy of predicting the path loss, i.e. (46.5% to 66.4%) of Hata model values.

KEYWORDS Propagation Model, Hata Model, Mobile Network, Cellular Network, Path Loss.

1 INTRODUCTION In wireless communication the information transfers from transmitter to receiver through wireless channel, which likely be affected by interference, fading, scattering, reflection …etc from the surrounded obstacles like buildings,

ISBN: 978-1-941968-12-3 ©2015 SDIWC

Dr. Tammam Benmusa EE Department, Faculty of Engineering University of Tripoli [email protected] terrains, cars, threes ..etc , due to all these processes the signal decays with distance[1]. The difference between signal strength transmitted from TX and signal strength received at RX defined as a pathloss, which is the main limitation of the cell coverage. The path loss is changeable and depending on many factors. One of the main important factor is the area type. The area can be classified as one of five types; Dense urban, Urban, Dense suburban, Suburban and Rural. The classification is based on the population density, building type and building density. The accuracy of predicted RSS is valuable in mobile network design, which is effected by cell characteristics (location- Antenna height– Antenna type- frequency band- transmitted power etc). Many works has been done to model this loss mathematically either in empirical or in theoretical ways. These models help engineers during the designing phase of the cellular networks. The empirical models is obtained from the measurement of the effect of environment conditions on the RSS, regardless the correlation between the environment parameters. This type of models is commonly used since it requires less executing time and much lower details of the environment conditions. In this work the RSS were measured as function of distance in 10 locations in Tripoli, 2 locations for each type of areas. For every location the measured values were averaged every distance equal to 40 λ. These values were analyzed by linear regression method to

5

The Proceedings of Second International Conference on Electrical and Electronics Engineering, Clean Energy and Green Computing, Konya, Turkey, 2015

develop the modified propagation model which is suitable for Tripoli Area[2],[3]. The rest of this paper is organized as follows. Section 2 gives a brief idea about the most famous propagation models, where section 3 describes the regression methods, and how to use linear regression method to get a mathematical equation that fits certain data. The measurements methodology used in this work has been illustrated in section 4, before presenting the results and the discussion in section 5 and 6. Finally, the paper was concluded in section 7. 2 PROPAGATION MODELS Propagation model is a mathematical tool to simulate the propagation loss that can be used by engineers and scientists in designing and optimizing the wireless networks. The main goal in the design phase of the wireless network is to predict the amount of the signal strength as a function of the separation distance between the transmitter and the receiver, which is affecting the cell coverage radius in cellular networks . It is also used to avoid the expected interference with the neighboring sites. Here are some examples of the propagation models. 2.1 Free Space Propagation Mode The free space model is used to predict the RSS when the transmitter and the receiver have clear, unobstructed line of sight path between them. The free space model predicate the path loss as function of frequency and the separation distance between the transmitter and the receiver as per the following equation [4],[7]; LdB = 32.44+20 log10(f)+20 log10(dkm) Where LdB is the free space path loss in dB f is the operating frequency in MHz dkm is the separating distance in km.

(1)

2.2 Lee propagation model W. Lee proposed a very simple signal propagation model originating from a series of measurements made in the USA at 900 MHz carrier frequency. According to the Lee model, the mean power measured at distance “d” from the transmit station is determined by [1],[4],[10]: (2) Where; L0 is the loss at 1km. υ is the loss parameter. αc is a correction factor. The prediction were done for a carrier frequency of 900MHz, a base station (BS) antenna of height 30.5, and a receiving antenna or mobile station (MS) height of 3 m. The correction factor αc is included to account for any change in the standard parameters used in the model and can be expressed as αc=10 log10(F0) F0 is the correction factor selected on the basis of a series of component factors according to the formula. (3) where the subsequent factors Fi are described by expressions (4) (5) The power υ =1 for the mobile station antenna height lower than 3m and υ =2 for the heights larger than 10m. (6) (7) F5 = GMS

(8)

Where; GMS is the MS antenna gain.

ISBN: 978-1-941968-12-3 ©2015 SDIWC

6

The Proceedings of Second International Conference on Electrical and Electronics Engineering, Clean Energy and Green Computing, Konya, Turkey, 2015

2.3 Hata Model The Hata model, also known as the Okumura– Hata model, is one of the most commonly used model for a macro cell environments to predict the radio signal attenuation. The model is considered as an empirical model, since it has been developed using field measurements data. The field measurements has been performed in Tokyo-Japan, and the obtained results was published in a graphical format and put into equations. The model is valid for quasi-smooth terrain in an urban area. For other terrain types correction factors are [5],[6],[7],[10],[11]. The ranges of the used parameters for this model are shown in table 1 Table 1. Hata Model Parameter Range Parameter Symbol Range Frequency range 150–1500 MHz f Frequency Extension 1500–2000 MHz Distance between MS and d 1–20 km BTS Transmitter antenna height Hb 3–200 m Receiver antenna height Hm 1–10 m

The Okumura–Hata model for path loss prediction in Urban area can be written as L = A + B log10 (f) - 13.82 log10 (Hb ) - a(Hm) + [44.9-6.55 log10 (Hb) ] log10 (d) (9) Where:f is the frequency (MHz). Hb is the base station antenna height (m). Hm is the MS antenna height (m). a(Hm) is the mobile antenna correction factor d is the distance between the BTS and MS (km). The correction factor for the MS antenna height is represented as follows, for a small or medium sized city: a(Hm) = [1.1 log10 (f) - 0.7] Hm - [1.56 log10 (f) - 0.8] (10) and for large city:

ISBN: 978-1-941968-12-3 ©2015 SDIWC

(11) The parameters A and B are dependent on the frequency as follows [1-4-5]:

For sub urban area L= L urban – 2[log10(f/28)]2 - 5.4

(12)

For rural areas L = L urban – 4.78[log10(f/28)]2 + 18.33 log10(f) 40.94 (13) Hata model is not suitable for micro-cell planning where antenna is below roof height. 3 LINEAR REGRESSION Regression analysis is a statistical technique for investigating and modeling the relationship between variables from previous statistics data. Applications of regression are numerous and occur in almost every field, including engineering, the physical and chemical sciences, economics, managements, biological sciences and the social sciences. In fact regression analysis may be the most widely used statistical technique[8]. There are three types of regression; linear regression, multi linear regression and nonlinear regression. The linear regression type has been used in this work because it is the most suitable type for such work, and will be discussed briefly below[9]. In a linear regression model, the variable of interest (the so-called “dependent” variable) is predicted from another variable (the so-called “independent” variables). A line representing the best fitting to the relation between the dependent and the independent variable, which are the scattered points in the x-y plan, is obtained. This line is called best fitting line,

7

The Proceedings of Second International Conference on Electrical and Electronics Engineering, Clean Energy and Green Computing, Konya, Turkey, 2015

and can be represented mathematically from equation 14. yi = β0 + β1xi + εi, i=1,2,…,n

(14)

Where: y denotes the dependent variable, x1, …, xn, are the independent variables, β0 called the intercept, is the predicted value of y when xi=0, β1 called the slop, may be intercepted as a change in y for unit change in x. εi, is representing the error, which is the difference between the actual data values and the best fit line calculated values. The Root Mean Square Errors (RMSE) method is used to obtain the values of the parameters β0 and β1 as follows[9]. Equation (14) can be rewritten as εi = yi - β0 - β1xi,

i=1,2,…,n

(15)

The summation of squared error will be (16) The value of are given by

0

and

1

The areas of Tripoli has been divided into small parts based on their morphology, which is depends on the population density, the height of buildings and how far they are separated from each others. Each part of Tripoli was classified as one of five area types; Dense Urban Urban- Dense Suburban - Suburban- Rural. In this work the classification of Tripoli parts was based on previous work done by Ericsson company[12]. The measurement environments were chosen to have high accuracy results. Transmitter parameter 1- Transmitter power is 20dBm. 2- The feeder loss is 38dBm. 3- Antenna gain is 2 dBi. 4- Frequency is 900MHz 5- Transmitter height is 28m in Dense urban area and 20m for the others. Receiver parameter 1- Antenna gain 2 dBi. 2- Receiver height is 1.5m.

that minimize 2

=Σ(yi- )(xi- )/Σ(xi- )

(17)

and (18) where: is the average of the dependent variable yi. is the average of the independent variable xi. The values of and are called the least squares estimates of and and will be used in the final form of the equation. The intercept and the slope of the line that has the smallest possible sum of squares of the vertical distances from each point to the line. For this reason, the line is also called the least squares regression line, and the fitted value will be given by; (19)

ISBN: 978-1-941968-12-3 ©2015 SDIWC

4 MEASUREMENTS METHODOLOGY

Ten locations were chosen to conduct the measurement in order to get ten different paths, two paths for each area type. The RSS measurement was taken in each path starting from the base station to about 1 km apart from the base station. The measurement rate was 15 sample for 40 λ, where λ is the wavelength of the measured signal. Each 15 samples were averaged and subtracted from the transmitted power to get the path loss corresponding to the average distance of these 15 samples. The values of these pathloss and the corresponding distances were put in a table. The above process was repeated for each path of the selected 10 paths and ten tables were obtained. Two tables for each area type. The linear regression method was applied to the values of the first table of each area type to obtain the best fitting line. Which is representing the correction factor in the

8

The Proceedings of Second International Conference on Electrical and Electronics Engineering, Clean Energy and Green Computing, Konya, Turkey, 2015

modified Hata model that matches the Tripoli environment. The values of the second table of each area type is used to verify the model results and to calculate its accuracy, by implementing the Root Mean Square Error method (RMSE). The RMSE was calculated between actual measured path loss value and the calculated values from the modified model using the following equation

LHata U is the Hata path loss for Urban area. b) For each area type, the values of the path losses obtained from the measurements were plotted as a function of distance along with the modified model in one graph. Also another line was drown in the same graph which is representing the Hata model values. The graphs for Dense urban, Urban, Dense suburban, Suburban and Rural area types are shown in figures 1, 2, 3, 4, and 5 respectively.

(20) Where; Pm: Measured path loss (dB) Pr: calculated path loss from the modified model (dB) N: Number of measured data points The Standard Deviation (STD) is also calculated for measurements. The accuracy assessment using RMSE was done for both Hata model and the modified model. Both results were put in one table for the sake of comparison.

Figure 1. Results for Dense Urban area.

5 OBTAINED RESULTS The results of this work has been presented as follows:a) The equations of the modified model is given also for each type of area as follows; Lmod DU = 7.32 + 0.649 LHata U

(21)

Lmod U = 17.5 + 0.655 LHata U

(22)

Figure 2. Results for urban area.

Lmod DSU=47.89+0.63(7.32+0.649LHataU) (23) Lmod SU =9.87+0.74(17.5 + 0.655 LHata U ) (24) Lmod R =32.82+0.69(17.5 + 0.655 LHata U )(25) Where; Lmod DU, Lmod U, Lmod DSU, Lmod SU , and Lmod R are the modified path loss for Dense urban, Urban, Dense suburban, Suburban and Rural area types respectively.

ISBN: 978-1-941968-12-3 ©2015 SDIWC

Figure 3. Results for Dense Suburban area.

9

The Proceedings of Second International Conference on Electrical and Electronics Engineering, Clean Energy and Green Computing, Konya, Turkey, 2015

Table 3.RMSE and STD utilizing second path (model validation) HATA model Modified HATA model RMSE STD RMSE STD DU 24.9438 9.251 9.7334 6.3422 U 20.7096 10.603 7.1384 5.79 DSU 6.8513 6.87 7.0395 7.1207 SU 20.9048 6.3814 5.4052 4.6939 R 10.4517 10.295 11.9927 4.5127

6 RESULTS DISCUSSION COMMENTS Figure 4. Results for Suburban area.

Figure 5. Results for Rural area.

c) The RMSE and the STD for both Hata and the modified Hata models were calculated for the all area types are presented in table 2. The RMSE used the pathloss values of the first path which are used in developing the modified model. Table 2 RMSE and STD utilizing first path (model accuracy- assessment) HATA model Modified HATA model RMSE STD RMSE STD DU 35.5029 5.0616 5.2647 4.677 U 20.3028 6.9172 5.4649 5.2946 DSU 10.1761 5.5055 5.3426 4.4787 SU 27.7714 7.0201 4.6089 4.6506 R 11.6074 8.4583 6.879 2.966

d) The modified model for each area type was validated by calculating the RMSE and STD from the measured values of the second path and the modified model obtained from the first path. These values are given table 3

General comments on the obtained results are given below in the following points:1- The RSME between the Hata model and the measured data is varying from 10.1 to 35.5 dB, where the RSME was within 4.6 to 6.8 dB between the modified model and the measured data in first path and 5.4 to 11.9 dB for the measured data for second path. That is mean the modified model gives 4.7 to 23.6 dB improvement in the accuracy of predicting the path loss (46.5% to 66.4%) 2- This improvement is expected since the defined urban area in Tokyo is totally different from the urban area in Tripoli. The difference is from the height, buildings density, and width of roads point of views. 3- Another cause for improvements is the environmental and topological differences between Tokyo and Tripoli. 4- the values of RMSE and STD for different areas in Tripoli whereas the acceptable range is up to 6 dB [11]. 7 CONCLUSION The accuracy of predicting the RSS is key issue in mobile network design. Several works have been done to the propagation path loss mathematically. Hata model, which has been devolved based on measurements done in Tokyo, is the most used one in designing cellular network because of its simplicity. A lot of work has been carried out to modify the model to get accurate results with other places. This paper illustrated the modification process of Hata model at frequency band [900]MHz to match the environments at the Tripoli area. The model is

ISBN: 978-1-941968-12-3 ©2015 SDIWC

10

The Proceedings of Second International Conference on Electrical and Electronics Engineering, Clean Energy and Green Computing, Konya, Turkey, 2015

depending on the area type, since there are 5 types of areas depending on their morphology.

The path loss was measured in ten different paths. Five of them were used to develop the modified model suitable for Tripoli averments using the linear regression method, which was used to adopt the Hata model to the measured values. Where the values of the other paths were used to verify the developed model using the root mean square error (RMSE) method, which is used to evaluate the new model accuracy. The devolved model showed 4.7 to 23.6 dB improvement in the accuracy of predicting the path loss, i.e. (46.5% to 66.4%) on the original Hata model.

REFERENCES [1] [2]

[3]

[4] [5]

[6] [7] [8] [9] [10] [11]

[12]

P.M. Shankar, Introduction to wireless system,1st Edition, John Wiley and sons,2002. E. Östlin, On Radio Wave Propagation Measurements and Modelling for Cellular Mobile Radio Networks, Blekinge Institute of Technology, 2009. J. Lempiäinen and M. Manninen, Radio interface system planning for GSM/GPRS/UMTS, Kluwer Academic Publishers,2001. J. S. Seybold, Introduction to RF propagation, 1st Edition, John Wiley and sons,2005. Rappaport, T. S, Wireless Communications Principles and Practices, 2nd Edition, Prentice Hall PTR, Upper Saddle River, NJ 2002. A. Mishra, Advanced celluler network planning and optimization, John Wiley and sons,2007. A. Goldsmith, WIRELESS COMMUNICATIONS, 2004. D.C. Montgomery, E. A. Peck, G. Vining, Introduction to Linear Regression Analysis,2012. X. Yan, Linear Regression Analysis: Theory and Computing, 2009. K. Wesolowski, Mobile communication systems,1st Edition, John Wiley and sons,1999. Z. Nadir, Empirical, Pathloss Characterization for Oman, Publication Year: 2012 ,Page(s): 133 – 137,IEEE conference publication. Tripoli 5 meter- City geo data package for TEMS cellplanner, 2010, Geodata team, [email protected].

ISBN: 978-1-941968-12-3 ©2015 SDIWC

11

The Proceedings of Second International Conference on Electrical and Electronics Engineering, Clean Energy and Green Computing, Konya, Turkey, 2015

A Survey on Some Currently Existing Intrusion Detection Systems for Mobile Ad Hoc Networks Mnar Saeed Alnaghes

Fayez Gebali

Department of Electrical and Computer Engineering

Computer Engineering

University of Victoria

University of Victoria

Victoria, Canada

Victoria, Canada

[email protected]

[email protected]

Abstract Mobile Ad-Hoc Network (MANET) is one of the most promising technologies that have applications in military, environmental, space exploration and forestry industry areas. This type of network has attractive features such as its low transmission power to conserve energy, increase throughput, and reduce delay. However, it suffers from many constraints, including limited resources, and the use of insecure wireless communication channels. Due to the lack of defense, the security of these networks is a worthy concern, particularly for the applications where confidentiality has prime importance. Thus, any kind of intrusions should be detected before attackers can harm the network in order to operate MANET in a secure way. In this article, we present a survey of the state-of-theart in Intrusion Detection Systems (IDSs) that are proposed for MANETs. This is followed by a comparison of each scheme along with their advantages and disadvantages. This survey is concluded by highlighting open research issues in the field. Keywords Security, Network security, Mobile Ad-Hoc Network, MANET, Intrusion Detection System, IDS.

1

Department of Electrical and

Introduction

Traditional wired networks are relatively more secure compared to wireless networks. Wireless ad hoc network integrates with multiple wireless systems such as wireless local area network (WLAN), wireless personal area network (WPAN), and wireless metropolitan area network (MAN), in order to improve its performance. It also provides communication between various devices (nodes) using a shared wireless channel. However, unlike more conventional wireless networks, nodes in ad-hoc networks communicate without the assistance of a fixed network infrastructure. It uses collaborative store-and-forward strategy to provide connectivity beyond transmit/receive range. Nodes within one an-

ISBN: 978-1-941968-12-3 ©2015 SDIWC

other’s radio range can communicate through wireless links, and, often these nodes are mobile, they dynamically change their locations. This type of networks is suited for situations where rapid network coverage is required or it is highly costly to deploy and manage a network infrastructure. Wireless ad-hoc networks are vulnerable to threats because the wireless medium is insecure, and, the domain of attacks is transient in nature as are the wireless networks themselves. Thus, it needs efficient routing protocol and design particular detection and strategies prevention techniques in order to secure this network. For the importance of securing such a network, there has been much work in intrusion detection system for traditional wired networks, but, it is a bit different for Ad hoc networks. The key reason is the differences of architectural features, most notably the lack of a fixed infrastructure. The lack of certified or trusted nodes and lack of centralized audit points, such as routers, switches, and gateways, makes it difficult to collect audit data for the entire network. Intrusion detection system, which monitors system activities and detects intrusions, must work with localized partial information. An IDS is generally used to complement other security mechanisms, and, it is a complex and difficult task due to Ad-hoc network’s features. The rest of the paper is organized as follows: In Section II, an introduction of IDSs is reviewed. Section III includes a brief discussion of IDSs proposed for MANETs, followed by their classifications in Section IV. Then, a detailed literature review on MANETs IDSs is provided along with comments on their prominent and lacking features. Finally, this paper is concluded by comparing existing approaches.

2

Intrusion Detection System In General

Intrusion is any attempts to compromise the integrity, confidentiality, or availability of a resource and an intrusion detection system (IDS) is a system which

12

The Proceedings of Second International Conference on Electrical and Electronics Engineering, Clean Energy and Green Computing, Konya, Turkey, 2015

detects such intrusions [2]. There are three main components of an IDS. First, data collection which is responsible for collecting data and preprocessing its tasks; data storage, data format, and sending data to the detection module. IDSs are able to use different sources for data as inputs to the system such as system logs, network packets, etc. [3]. Second, detection, at this level, data is analyzed to detect intrusions. Third, response where the action will be taken after receiving indications of detected intrusions from detection component. There are three basic intrusion detection techniques as discussed in the following subsection [18]:

2.1

Anomaly-based intrusion detection:

The parameters of the system such as CPU usage for programs are profiled. Intrusions are recognized as variations from the normal behaviors. The challenge is to define the normal behaviors. A block diagram of a typical anomaly detection system is shown in Figure 1.After monitoring system activity and profiling it, the patterns are saved in the system profile where they can be updated. The overhead of updating several system profile metrics make anomaly IDS computationally expensive. Additonally, false positive can be high in anomaly-based detection. On the other hand, it is capable of detecting previously unknown attacks.

Figure 1: a typical anaomaly detection system

2.2

The Misuse-based or signaturebased intrusion detection:

Known attack signatures are compared with current system activities. This system is efficient and has a low false positive rate thus it is preferred by commercial IDSs. However, Misuse-based intrusion detection cannot detect new attacks. This needs continuous updating for new types of attacks. A block diagram of a typical misuse detection system is shown in Figure 2 below. Detection rules are generated based on specific attack’s signatures. These rules are saved and updated in the system profile. The main issues here are how to write a signature that encompasses all possible variations of the pertinent attack, and how to write signatures that do not also match non-intrusive activities.

Figure 2: a typical misuse detection system

ISBN: 978-1-941968-12-3 ©2015 SDIWC

2.3

Specification-based intrusion detection:

There is a specified set of constraints on a program or a protocol, and, intrusions are detected as runtime violations of these specifications. This approach combines the strengths of anomaly-based and misuse-based detection techniques. Thus, it can detect known and unknown attacks with a lower false positive rate. It can discover a new attack that does not follow the system specifications. Further, specification-based intrusion detection does not lead to false alarms when the program or protocol has unusual but legitimate actions, since it uses the legitimate behaviors of the program or protocol [4]. Also, this approach cannot detect some kind of attacks such as Denial of Service (DoS) attacks since these do not violate program specifications directly.

3

Intrusion Detection System for MANET

A fairly usable analysis of MANET’s risks by identifying assets, threats, and vulnerabilities will be helpful to provide good security solutions [7]. Besides, MANET’s features should be considered when a new IDS is being designed for this network. Without a centralized authority, the algorithms used for intrusion detection must be distributed in nature, yet it must be kept in mind that attacks may be made from nodes inside the network, this means that node which is participating in a collaborative intrusion detection algorithm may be a malevolent node. Further, nodes of a mobile ad hoc network can only see the packets that they send or receive together within their radio range. Furthermore, because of the wireless ad hoc network is distributed and cooperative, the intrusion detection system in MANET should also be distributed and cooperative, which shows certain difficulties. The highly dynamic feature of this network makes intrusion detection system’s traditional techniques become unreliable. The IDS needs a scalable architecture to collect sufficient evidences to detect the attacks effectively. Mobile nodes typically consume battery power, and, they have different capacities. These devices are varied, and their computational and storage capacities vary too. This variety affects the efficiency of IDS agents they support. For instance, nodes may drop packets to conserve resources causing difficulties in distinguishing failed or selfish nodes from attacker. Thus, it is very important for a detection algorithm to be considered the limited resources problem. Besides, Ad Hoc Networks usually have highly cooperative routing protocols which makes them the target of new attacks. MANE,Ts have more constrained bandwidth, and, the link breakages are common. An IDS agent needs to communicate with other IDS agents with a view to gain data or alerts and needs to be aware of wireless links. Due to the heavy traffic of intrusion detection systems, congestion could be caused. Intrusion detection systems agents will need to decrease their data transfers [1].

13

The Proceedings of Second International Conference on Electrical and Electronics Engineering, Clean Energy and Green Computing, Konya, Turkey, 2015

4

Intrusion Detection Classification

System

The existing IDS architectures for MANET consider under three basic categories [9]: (a) Stand-alone architecture which uses an engine installed at each node utilizing only the node’s local audit data. This fact limits the Stand-alone IDS in terms of detection accuracy and the type of attacks that it could detect [14].

a more realistic model, the dynamic game, allows the defender to update his belief about his opponent’s type based on new observed actions and the game history. Additionally, a Bayesian hybrid detection approach in the dynamic game in which a lightweight monitoring system is used for the defender, as it shown in Figure 3, to estimate his opponent’s actions, and a heavyweight monitoring system acts as a last resort of defense. The simulation’s results show that by adding the very simple coarse-grained node-to-node analysis system in front of the association-rule analysis system will reduce the probability of false alarm for the overall equivalent IDS.

(b) Cooperative architecture which processes each host’s audit data locally. It is similar to standalone architecture, but it also uses collaborative techniques to detect more accurately a large list of attacks. Thus, the majority of the most recent IDS for MANET are based on them and the Hierarchical architectures as well [14]. (c) Hierarchical architecture which is an extended version of the cooperative architectures [14]. This architectures reach to a multilayer approach by dividing the network into clusters. Specific nodes are selected as cluster-heads and undertake various responsibilities and roles in intrusion detection, which are different from those of the simple cluster members. (d) Intrusion Detection System architecture with mobile agent which uses mobile agents that can move through large networks. It allows the distribution of the intrusion detection tasks. Each agent is assigned to do a specific task and then one or more agents are distributed into each node in the network.

5

Existing Intrusion Detection System for MANET

As it is clear, it is difficult to build a completely secure MANET system in spite of using a complex cryptographic technique or so-called secured routing protocols. Some of the existing IDS algorithms that have been introduced for MANETs are explained as follows:

5.1

Bayesian Game Approach IDS [6]

A game theoretic framework using a Bayesian formulation which can analyze the interactions between pairs of attacking and defending nodes was introduced in a flat Ad-hoc network. The achievable equilibrium strategies was investigated for the attacker and defender game in both static and dynamic scenarios. In these games, players try to maximize their payoff. The motivation behind the Bayesian game formulation is that generally an attacker/defender game is an incomplete information game where the defender is uncertain about the type of his opponent whether it’s a regular node or malicious. A Bayesian game formulation provides a framework for the defender to select his strategies based on his belief on the type of his opponent[13]. In the proposed static game, the defender always assume fixed prior probabilities about the types of his opponent throughout the entire game period. On the other hand,

ISBN: 978-1-941968-12-3 ©2015 SDIWC

Figure 3: The Bayesian hybrid detection framework

5.2

Acknowledgment-Based IDS[10]

Approach

The 2-ACK scheme forwards two hop acknowledgment packets in the opposite direction of the routing path. This approach is an add-on method for routing schemes to detect and mitigate the effect of such routing misbehavior. The 2-ACK transmission is carried out for every set of three nodes along with the path from the source to the destination. Figure 4 shows the operation of the 2ACK scheme, N1, N2, and N3 are three consecutive nodes along a route, S and D are the source and destination nodes. By acknowledging every data packet transmitted over every three successive nodes along the route, 2ACK is able to predict malicious links. This scheme overcomes several problems including ambiguous collisions, receiver collisions, and limited transmission powers comparing to some other schemes. The simulation shows that the 2ACK scheme maintains up to 91 percent packet delivery ratio even when there are 40 percent misbehaving nodes in the MANETs. The regular Dynamic Source Routing (DSR) scheme is able only to offer a packet delivery ratio of 40 percent.

Figure 4: 2-ACK Scheme.

14

The Proceedings of Second International Conference on Electrical and Electronics Engineering, Clean Energy and Green Computing, Konya, Turkey, 2015

5.3

Ex-Watchdog Approach IDS [11]

The proposed intrusion detection system is an extension of Watchdog so the used system is the Routeguard as the response system, and the IDS function is also detecting intrusion from malicious nodes and reports this information to the response system. The authors declared this technique in view to overcome one of the weaknesses in watchdog which is the false misbehaviour node. In the simulation, the metrics of throughput and overhead were used in order to evaluate the ExWatchdog performance with some malicious nodes that falsely report other nodes as misbehaving. The benefit of their technique is that it decreases the overhead comparing with Watchdog, however, it does not increase the throughput obviously.

5.4

Classification-Based Approach IDS [12]

In this paper, intrusion detection system models using supervised classification algorithms are presented. The used classification algorithms are Multi-Layer Perceptron (MLP), the linear classifier, the Gaussian Mixture Model (GMM), the Naive Bayes and Support Vector Machines (SVM). They assume that the IDS architecture composed of multiple local IDS agents that are responsible for detecting possible intrusions locally. They used features from the network layer and evaluated the performance of these algorithms for the detection of four serious attacks in MANETs, the Black hole, Forging, Packet Dropping and Flooding attack. In results, the easiest attack to be detected is the Flooding attack, while the most difficult attack to detect is the Packet Dropping attack. The most efficient classifier for detecting all four types of attacks simultaneously is the SVM classifier for multiclass classification although the MLP classifier presents a satisfying Detection Rate (DR) and also a quite high False Alarm (FA) rate. Further, the highest the number of malicious node in the network the easiest to detect intrusions.

5.5

Zone-Based Approach IDS [8]

The proposed intrusion detection system is a nonoverlapping zone-based framework. The use of different detection techniques is pliable in their IDS agents, but technically they only use Markov chain anomaly detection in their research. Based on geographic partitioning, the network is divided into zones in order to save bandwidth. The performance is improved by obtaining data from many nodes. An intra-zone node is a regular node in a zone, and an inter-zone node is a gateway node which work as a bridge to other zones. Each node in the zone is responsible for local detection and sending alerts to the inter-zone nodes. The alert is generated and sent by inter-zone nodes based on the combined information received. The detection performance of the algorithm could be decreased with the increasing of the number of attackers. From simulation, the local anomaly detection model works well in low mobility environment. However, it is shown that the anomaly-based detection performs poorly because the data are unstable under high mobility. The benefits of the aggregation algorithm

ISBN: 978-1-941968-12-3 ©2015 SDIWC

generally are lower false positive and higher detection rate than local IDS achieves. One of the contributions is MANET Intrusion Detection Message Exchange Format (MIDMEF) which declares information exchange’s format between IDS agents.

5.6

Specification-based Approach IDS [4]

This paper introduces the first application of specification-based detection technique for MANETs. It is assumed that all nodes are covered using network monitors which have nodes IP addresses. All these monitors and their messages are secure, and, MAC addresses are not rigged. Each Network Monitor uses finite state machines as operations specifications of AODV in order to correct AODV routing behavior and distributed network monitors for detecting run-time violation. All RREQ and RREP messages in the range of the network monitors are monitored in request reply flows. The algorithm uses a tree structure of data which can detect most of known and unknown attacks against routing protocols efficiently in real time and with minimum overhead since these attacks have clearly defined specifications. However, network services can be affected by dropping certain broadcast messages if this action is happened at a critical point. Also, according to MANET characteristics, nodes can independently join and leave network and frequently move, it is non pragmatic to assume that network monitors can cover all nodes and have all their IP and MAC addresses which are non-forged.

5.7

Fuzzy Logic Approach IDS [15]

The proposed intrusion detection system uses fuzzy logic, this fuzzy logic works to handle imprecise information in order to help the IDS to detect malicious behavior and identify the attacks. Fuzzy logic is a computational paradigm that provides a mathematical tool to deal with the uncertainty and the imprecision which is involved in human reasoning. Fuzzy logic’s features, which is the capability to express knowledge in a linguistic way, makes fuzzy logic-based systems attractive for applications like medical diagnosis. After detecting intrusion in MANET, fuzzy Logic approach is used for three types of symptoms. First symptom, when the number of packets dropped is greater than threshold. Second symptom, when the number of packets dropped is greater than threshold and the number of packets dropped for the particular destination is greater than another threshold value which called DestThreshold. Third symptom, when the number of packets drop is greater than threshold and the number of packets drop for the particular source is greater than DestThresold. The main aim behind this method is to detect the malicious nodes locally at each node. Thus, each node has its local IDS which monitors the nodes traffic. From simulation results, the output of a fuzzy-based IDS could be a more accurate input since the output of an IDS is normally used as input to a routing protocol for isolating malicious nodes from the network. Additionally, analyzing the level of intrusion in the network using the fuzzy logic is more accurate than when it is not used.

15

The Proceedings of Second International Conference on Electrical and Electronics Engineering, Clean Energy and Green Computing, Konya, Turkey, 2015

5.8

Elliptic Curve Cryptography-Based Enhanced Adaptive Acknowledgment IDS [16]

An acknowledgment-based IDS named Elliptic Curve Cryptography Based Enhanced Adaptive Acknowledgment (ECC-EAACK) is proposed. It demonstrates higher rates for malicious behavior detection in certain situations while does not greatly affect the network performances. It has a strong and light-weight Intrusion detection mechanism called EAACK which requires less hardware cost. Acknowledgment (ACK), secure ACK (S-ACK) and misbehavior report authentication (MRA) are the three major parts of EAACK. All the three parts rely on acknowledgment packets to detect misbehaviors in the network. Thus, it is extremely important to ensure that all acknowledgment packets in EAACK are authentic and untainted. ACK is basically an end-to-end acknowledgment scheme, which acts as a part of the hybrid scheme in EAACK in order to reduce network overhead when no network misbehavior is detected. If the receiver node does not send the ACK within predefined period, then ACK assumes malicious may attend, as a consequence, it switch to S-ACK part to detect malicious nodes. In S-ACK, for every three successive nodes in the route, the third one is required to send an S-ACK acknowledgment packet to the first node. The objective of starting S-ACK mode is to detect misbehaving nodes in the presence of receiver collision or limited transmission power. When a malicious node is found, then MRA part suggests alternate path to the destination. The source node will not immediately trusts the misbehavior report because EAACK requires the source node to switch to MRA node and confirm this misbehavior report. This is a vital step to detect false misbehavior report in their proposed scheme. An improved Cryptographic technique was used in view to improve intrusion detection results in MANET, ensure the secure communication of packets, Increase security and convenience, and provide digital signatures that cannot be forged. After simulation, the results show positive performances which make this approach suitable to be implemented in MANETs.

5.9

Cross Layer-Based Approach for IDS [17]

In this paper, a cross layer-based detection system is presented to detect the black-hole attack in MANETs. This cross layer technique incorporating IDS leads to an escalating detection rate in the number of malicious behavior of nodes increasing the true positive and reducing false positives in the MANET. In this approach, the decision to detect malicious nodes is made by the network layer, MAC layer and physical layer on the basis of the parameters passed by the physical and the link layer to the network layer using cross layer design, and then the best path is selected among the various paths between the source node and the destination node. Each layers is designed separately with the services it implements, with the help of the well-defined interfaces through which these layers communicate with each other. In the layered architecture, UDP packets are sent to and from the network layer to the application layer

ISBN: 978-1-941968-12-3 ©2015 SDIWC

over the transport layer. This communication causes some delay which decreases the overall network performance. If it is possible to design a direct application layer- network layer interface bypassing the transport layer, the end to end delay is saved and hence the overall network performance is improved. Performance can be optimized by using the available information across many protocol layers. One of the vital methods of cross layer design is sharing of the database between the different layers hence the parameters could be available at different layers of the protocol stack. The nodes were simulated with random mobility model, and, a black hole attack was simulated with malicious nodes. As a simulation result, the throughput, which is the number of successfully delivered packets from the source to destination, and packet delivery ratio which Ratio, which is the parameter which tells how successfully the packets are delivered to the destination, of the proposed technique shows better results than that of the standard AODV under attack.

6

Comparison of Existing IDSs’ approaches

We have followed a critical analytical comparison to compare the discussed intrusion detection techniques since they have different routing protocols. Unluckily, these approaches are not entirely ideal. However, Specification-based Approach IDS [4] would be considered in terms of reducing the overhead rate and covering many of different routing attacks. The findings are summarized in Table 1.

7

Conclusion

Since mobile ad hoc networks has become widely used, the importance of securing these network has increased. Due to prevention techniques alone are not sufficient and new intrusions continually attack such a network, an IDS is a necessary component of a security system. IDS is basically used to detect possible violations of a security policy by monitoring system activities. The IDS perform some tasks which are monitoring the network, analyzing collected data, identifying intruders, generating alarms, and tracking intruders to prevent such attacks in the future. All these functionalities are in order to identify either an outside intrusion or an inside intrusion. Designing a powerful IDS that support the network performance is in fact a complicated task because of the MANETs Features. In this survey paper, we present brief details of various available intrusion detection techniques. The basic idea behind all of the above discussed IDSs is comparing their techniques in order to cope up with the limitations of IDSs. It is noticed that almost all intrusion detection systems are distributed and have a cooperative architecture. And, the specification detection is more powerful among the various detection methods used. However, IDS itself may be attacked and the study of the defense to such attacks should also be explored.

16

Bayesian Game Approach [6]

Acknowledgment-Based

1

2

ISBN: 978-1-941968-12-3 ©2015 SDIWC

9

Elliptic Curve Cryptography

8

Approach [17]

Cross Layer Based

Based Approach [16]

Fuzzy Logic Approach [15]

7

Techniques

Authentication

Mechanisms

Encryption

Scheme

Fuzzy Logic

Monitors

Network

Scheme

Markov Chain

Algorithms

Classification

Supervised

Mechanisms

Encryption

Misuse

Specification

Specification

Specification

Anomaly

Anomaly

Specification

Specification

Anomaly

Technique

AODV

EAACK

AODV

AODV

Modified

DSR

Unidentified

DSR

Unidentified

Black hole Attack

dropping

the middle, and, packet

Denial of Service, Man in

Gray hole

Black hole and

protocols

against routing

unknown attacks

Most known and

Routing Disruption

and,Flooding

Black hole, Forging,

Routing Disruption

Selfish, Sneak, and

Disruption

Selfish and Routing

Sneak Attack

Attacks

Protocol Unidentified

Type of

Routing

Table 1: Intrusion Detection Systems’ Comparison

Hierarchical

Cooperative

Cooperative

Cooperative

Specification Based

Approach [4]

Cooperative

Stand-alone

Cooperative

Scheme

2-ACK

Scheme

Hierarchical Cooperative

Game Theoretic

Method

Cooperative and

Architecture

Zone Based Approach [8]

6

5

Classification Based

4

Approach [12]

Ex-Watchdog Approach [11]

3

Approach [10]

IDS

No.

YES

YES

NO

NO

YES

YES

NO

YES

YES

Overhead

The Proceedings of Second International Conference on Electrical and Electronics Engineering, Clean Energy and Green Computing, Konya, Turkey, 2015

17

The Proceedings of Second International Conference on Electrical and Electronics Engineering, Clean Energy and Green Computing, Konya, Turkey, 2015

[12] Mitrokotsa, A., Tsagkaris M., and Douligeris,

References

Ch, Intrusion Detection in Mobile Ad Hoc Net[1] Smith AB, An Examination of an Intrusion Detection Architecture for Wireless Ad Hoc Net-

works Using Classification Algorithms, Boston: Spring, 2008.

works, 2001. [13] H. Otrok, N. Mohammed, L. Wang, M. Deb[2] Uppuluri

P,

Sekar

R,

Experiences

with

Specification-based Intrusion Detection, 2001. [3] Lundin E., Jonsson E., Survey of Intrusion Detection Research, 2002. [4] Tseng C-Y, Balasubramayan P, A SpecificationBased Intrusion Detection System for AODV, 2003. [5] Mishra, A., Nadkarni, K., Patcha, A., Intrusion Detection in Wireless Ad Hoc Networks. IEEE Wireless Communications, 2004. [6] Yu L., A Bayesian Game Approach for Intru-

babi, and P. Bhattacharya, A game-theoretic intrusion detection model for mobile ad hoc networks, Comput. Commun., 2008. [14] S. Sen, J. A. Clark, Intrusion Detection in Mobile ad hoc networks, in Chapter 17, Guide to Wireless Ad Hoc Networks, Springer, 2009. [15] Monita W. and Ningrinla M., Intrusion Detection in MANET using Fuzzy Logic, In IEEE 3rd National Conference, March 2012. [16] C Deboral, TIC

CURVE

G. Sudhagar,

AN ELLIP-

CRYPTOGRAPHY

BASED

sion Detection in Wireless Ad Hoc Networks,

ENHANCED ADAPTIVE ACKNOWLEDG-

GameNets06, October14, 2006.

MENT (ECC-EAACK) INTRUSION DETEC-

[7] Antonio M., A Platform Independent Risk

TION SYSTEM, International Journal of Com-

Analysis for Mobile Ad hoc Networks, Proc. of

puter Science and Mobile Computing, February

the Boston Univ. Conference on Information

2014.

Assurance and Cyber Security, 2006.

[17] S. Sharma, R. Mishra, A Cross Layer Ap-

[8] Sun B, Wu K, Zone-Based Intrusion Detection

proach for Intrusion Detection in MANETs, In-

System for Mobile Ad Hoc Networks, Int J of

ternational Journal of Computer Applications,

Ad Hoc and Sens Wirel Netw, 2006.

2014.

[9] Tiranuch A., Jie W., A survey on Intrusion Detection in Mobile ad Hoc Networks, Book Series

[18] Aurobindo S. , An Introduction to Intrusion Detection, www.acm.org ,2001.

Wireless Network Security, 2006. [10] Kejun L., Jing D., Varshney P.K., Balakrishnan K., An acknowledgment -based approach for the detection of routing Misbehaviour in MANETs, IEEE Trans. Mobile Comput., May 2007. [11] N. Nasser and Y. Chen, Enhanced intrusion detection systems for discovering malicious nodes in mobile ad hoc network, In Proc. IEEE Int. Conf. Commun. Glasgow, Scotland. 2007.

ISBN: 978-1-941968-12-3 ©2015 SDIWC

18

The Proceedings of Second International Conference on Electrical and Electronics Engineering, Clean Energy and Green Computing, Konya, Turkey, 2015

Harmonics Mitigation and Power Quality Improvement in Grid Connected Wind Power Plant Using STATCOM Nasiru B. Kadandani

Department of Electrical Engineering, Bayero University, Kano, Nigeria [email protected] Abstract—Grid integration of intermittent renewable energy sources like wind and solar is usually associated with power quality disturbance especially when the penetration level is high. This paper investigates the power quality issues associated with grid connected wind power plant (WPP) and how to overcome such problems using static synchronous compensator (STATCOM) – a member of Flexible A.C Transmission System (FACTS) devices. The major power quality issue investigated in the paper is harmonic distortions. Other power quality parameters such as voltage dip, voltage swell, flicker, transient, short and long interruptions were also briefly introduced. The entire system including the STATCOM control scheme is simulated in Simulink environment of MATLAB. Simulation results show that the harmonic contents at the output of the system with unfiltered linear loads are quite high, thus showing power quality deterioration. Fast Fourier Transform (FFT) analysis on the voltage and current waveforms of the system show that the STATCOM control scheme is capable of cancelling out the reactive and harmonic parts of the load current thereby restoring the smooth sinusoidal waveform of the voltage and current at the wind point of interconnection (POI), or point of common coupling (PCC) which signifies power quality improvement. Keywords—power quality; STATCOM; THD; FFT; WPP.

I. INTRODCUTION Power quality is a term used by engineers to refer to the problems associated with voltage, current or frequency deviation that result in failure or equipment malfunction. It is usually related to maintaining a smooth sinusoidal waveform of bus voltage, current and frequency [1]. Thus, a perfect power supply is one that is highly available within prescribed voltage and frequency tolerances, with smooth sinusoidal wave form and also enables equipment work properly. On the other hand, any disturbance in the magnitude, frequency or purity from the balanced three phase sinusoidal wave form can lead to deterioration of the power quality. Power quality is of great concern to consumers, utilities and equipment manufacturers. It becomes especially important with the insertion of sophisticated devices, whose performance is very sensitive to the quality of power supply [2]. Power quality study is gaining much emphasis due to the increased proliferation of sensitive equipment and the need of maintaining excellence in electrical power generation [3]. Increased research in grid integration of wind turbine and the

ISBN: 978-1-941968-12-3 ©2015 SDIWC

actual project development makes power quality issues more and more important both in theory and practice. Among all forms of power quality issues in grid connected wind power plants (WPP), harmonic distortion is more common since any deviation from a perfect sinusoidal wave shape in voltage, current or frequency is an indication of the presence of harmonics in the power system. Due to the wide-spread use of power electronic devices in residential, commercial and industrial loads, electric utilities have experienced an increase in the level of harmonic frequencies on their electrical delivery systems. Hence, the potential harmonic effects on power equipment and system operation are becoming a concern for the utilities. In response to the aforementioned challenges, some utility operators and wind farm vendors usually assess the quality of power supplied to major customers through measurement and analysis to ascertain whether the harmonic levels are within allowable limits as recommended by regulating bodies like IEC and IEEE. Fortunately, intensive research in power electronics has lead to the development of flexible a.c transmission system (FACTS) devices for power quality improvement. Such devices are capable of mitigating power quality disturbances with respect to grid codes and provide power system stability by controlling the power flow in a transmission network using power electronic controlled devices [4]. FACTS devices are categorized into two, namely; Thyristor-based FACTS controllers and GTO based FACTS controllers. Members of the first category include Static VAR Compensator (SVC), Thyristor Controlled Series Capacitor (TCSC), Thyristor Controlled Phase Angle Regulator (TCPAR), e.t.c. The second category includes static synchronous compensator (STATCOM), Static Synchronous Series Compensator (SSS), Unified Power Flow Controller (UPFC), e.t.c. II. POWER QUALITY PARAMETERS AND THEIR CONSEQUENCES A. Harmonic Distortions Harmonic distortion refers to the periodic deviation of voltage (or current) sine wave from a smooth sinusoidal shape and usually occurs when frequencies of the multiple integers of the fundamental frequency are added to the pure sinusoidal waveform of the voltage (or current). Harmonics are therefore

19

The Proceedings of Second International Conference on Electrical and Electronics Engineering, Clean Energy and Green Computing, Konya, Turkey, 2015

integer multiples of the fundamental frequency. Total harmonic distortion (THD) is a measure of all the values of the waveform that are distorted. At the generation end, harmonics are usually caused by power electronic converters used in variable speed wind turbine, while at the receiving (consumer) end, harmonics are caused by non-linear loads (equipments that do not use voltage or current in normal sinusoidal shape) such as computers, television, arc furnace, arc welders, mercury lamps, electronic ballasts, battery charger, variable speed drives, medical diagnostic equipment and fluorescent lamps [5]. Other causes of harmonics include resonance phenomena, transformer saturation, light dimmers and switch mode power supplies. Harmonics can cause significant damage to transmission and distribution network as well as to consumer equipment. The most notable effects of harmonic distortion are; overload operation of electric motor, nuisance tripping of circuit breaker, misbehaved operation of fuses, tripping of variable speed drives, overheating of electric motors and transformers, incorrect power measurement and failure or damage to electrical equipments like contactors and power factor correction capacitors. Harmonic distortions can be mitigated by application of active conditioners, input chokes, passive filters, isolation transformers, and equipments with build-in power factor correction capability such as FACTS devices like STATCOM, DVR, SVC. •

Harmonic Distortion in Bus Voltages

The total harmonic distortion of bus voltage, 𝑉𝑇𝐻𝐷 is defined as the ratio of the square root of the sum of squares of the rms value of the harmonic component, 𝑉𝑛 to the rms value of the fundamental vomponent, 𝑉1 [6]. 𝑉𝑇𝐻𝐷 = �∑40 ℎ=2

𝑉𝑛2 𝑉1

(1)

100

For 132kV, the voltage THD limit is